HAL Id: tel-02972475
https://hal.archives-ouvertes.fr/tel-02972475
Submitted on 20 Oct 2020HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Generation Networks
Arsenia Chorti
To cite this version:
Arsenia Chorti. Security Protocols and Resource Allocation for Fifth Generation Networks. Signal and Image processing. CY Cergy Paris Université, 2020. �tel-02972475�
Security Protocols and Resource Allocation for Fifth
Generation Networks
Arsenia (Ersi) Chorti
Maˆıtre de Conf´erences `a l’ENSEAHabilitation `a Diriger des Recherches de CY Cergy Paris Universit´e Section CNU 27, Informatique
Defended on the 12th October 2020 in front of the jury composed of Jean-Marie GORCE Professor, INSA-Lyon, President of the jury Gerhard FETTWEIS, Professor, TU Dresden, Vodafone Chair, Reviewer Marceau COUPECHOUX, Professor, T´el´ecom ParisTech, LTCI, Reviewer
Ghaya REKAYA, Professor, T´el´ecom ParisTech, LTCI, Reviewer
Camilla HOLLANTI, Professor, Aalto University, Dep. of Mathematics and System Analysis, Examiner Inbar FIJALKOW, Professor, ENSEA HDR Guarantor
and their theses’s directors, unless otherwise stated. Information derived from the published and unpublished work of others has been acknowledged in the text and references are given in the list of
sources.
Arsenia (Ersi) Chorti (2020)
The copyright of this thesis rests with the author and is made available under a Creative Commons Attribution Non-Commercial No Derivatives licence. Researchers are free to copy, distribute or transmit the thesis on the condition that they attribute it, that they do not use it for commercial purposes and that they do not alter, transform or build upon it. For any reuse or redistribution, researchers must make clear to others the licence terms of this work.
promising and far fetching – whenever possible – research topics in the general framework of fifth generation (5G) wireless. The works presented in this thesis reflect our studies in two areas of central importance for bringing 5G to life: wireless security and resource allocation.
With respect to security, novel challenges emerged in 5G with the Internet of things (IoT) paradigm and device to device (D2D) low latency communications. Novel verticals, such as haptics and vehicle to everything (V2X), require low complexity and low latency security mechanisms, particularly in the context of device authentication. In the present manuscript, lightweight solutions for device authentication using physical unclonable functions (PUF) and secret key generation (SKG) at the physical layer are presented.
Furthermore, as video content is responsible for more than 70% of the global IP traffic, it is important for content delivery infrastructures to rapidly detect and respond to changes in content popularity dynamics. In this thesis, we propose a flexible edge resource allocation approach leveraging unikernel and container technologies. The allocation of the edge server resources is driven by a real-time and low-complexity content popularity detector, implemented using off-line and on-line change point analysis. Variations of these algorithms have applications in intrusion detection in wireless sensor software defined networks, discussed next.
Finally, the potential use of non-orthogonal multiple access (NOMA) in the wireless uplink is considered. Early results on the performance comparison of NOMA vs orthogonal allocation schemes in asymptotic regimes, show that the gains in using NOMA carry on to the scenario of communications under statistical delay quality of service (QoS) constrains.
Dans mon rˆole de co-encadrent de th`ese, je me suis efforc´e de donner `a mes ´etudiants l’occasion de travailler sur des sujets de recherche prometteurs et fondamentaux dans le cadre g´en´eral de communica-tions sans fil de cinqui`eme g´en´eration (5G). Les œuvres pr´esent´ees dans cette th`ese refl`etent nos ´etudes dans deux domaines d’importance centrale pour la r´ealisation de la 5G : la s´ecurit´e et l’allocation des ressources.
En ce qui concerne la s´ecurit´e, de nouveaux d´efis sont apparus en 5G avec le paradigme de l’Internet des objets (IoT) et les communications device to device (D2D) `a faible latence. Les nouvelles verticales, telles que l’haptique et les communications v´ehiculaires (V2X), n´ecessitent une faible complexit´e et des m´ecanismes de s´ecurit´e `a faible latence, en particulier dans le contexte de l’authentification. Dans cette th`ese, des solutions d’authentification de l´eg`ere complexit´e en utilisant des fonctions physiques inclonables (PUF) et des g´en´erations de cl´es secr`etes (SKG) `a la couche physique sont pr´esent´ees.
En outre, comme le contenu vid´eo est responsable de plus de 70% du trafic IP mondial, il est important que les infrastructures de diffusion de contenu d´etectent et r´epondent rapidement aux changements de la dynamique de popularit´e du contenu. Dans cette th`ese, nous proposons une approche flexible d’allocation des ressources qui tire parti des technologies unikernel et containers. L’allocation des ressources est entraˆın´ee par un d´etecteur de popularit´e de contenu en temps r´eel et `a faible complexit´e, mis en œuvre `a l’aide des analyses hors ligne et en ligne des points de changement. Des variantes de ces algorithmes ont des applications dans la d´etection d’intrusion dans les r´eseaux d´efinis par les logiciels de capteurs sans fil, qui sont discut´es ensuite.
Enfin, l’utilisation potentielle d’un acc`es multiple non orthogonal (NOMA) dans le lien ascendant sans fil est envisag´ee. Les premiers r´esultats de la comparaison des syst`emes d’allocation NOMA par rapport aux sch´emas orthogonaux dans les r´egimes asymptotiques, montrent que les gains dans l’utilisation de NOMA se poursuivent dans le sc´enario des communications sous des contraintes statistiques de d´elai de qualit´e de service (QoS).
I would like to take this opportunity to thank wholeheartedly the Telecom girls, Inbar Fijalkow, Iryna Andriyanova, Veronica Belmega, Marwa Chafii, Laura Luzzi, and, also Myl`ene Pischella, Marine
Abstract 3
Acknowledgements 4
Nomenclature 13
1 Activity Review 16
1.1 Motivation for Application for the HdR Diploma . . . 16
1.2 Curriculum Vitae . . . 17
1.3 Publication List . . . 23
1.3.1 Books [B] / Book Chapters [BC] . . . 23
1.3.2 Refereed International Journals [J] . . . 23
1.3.3 Refereed International Conference Proceedings [C] . . . 24
1.3.4 Posters . . . 27
1.3.5 In Preparation [U] / Submitted [S] . . . 27
1.4 Recent Research Results . . . 28
1.4.1 Motivation on Studying Physical Layer Security and Resource Allocation for 5G Systems . . . 28
1.4.2 Results in Resource Allocation . . . 29
1.4.3 Results in PLS . . . 30
1.5 Recent Teaching Activities . . . 33
1.5.1 Overview of Teaching Activities in France (ENSEA) . . . 33
1.5.2 Overview of Teaching Activities in the UK . . . 34
1.6 Research Supervision . . . 36
1.6.1 PhD Theses to be Defended in September 2020 . . . 36
1.6.2 Ongoing Theses . . . 37
1.6.3 Current Postdoctoral Students . . . 37
1.7 Structure of the Rest of the Thesis . . . 37
References . . . 38
2 Security Protocols for Internet of Things Applications 39 2.1 Introduction . . . 39
2.2 Contributions and Chapter Organization . . . 39
2.2.1 Threat Model . . . 41
2.2.2 Notation . . . 41
2.2.3 Chapter Organization . . . 41
2.3 Related Work . . . 41
2.4 Node Authentication Using PUFs and SKG . . . 42
2.4.1 Node Authentication Using PUFs . . . 43
2.4.2 SKG Procedure . . . 43
2.5.1 Parallel Approach . . . 50
2.5.2 Sequential Approach . . . 52
2.6 Effective Data Rate Taking into Account Statistical Delay QoS Requirements . . . 53
2.7 Results and Discussion . . . 56
2.7.1 Numerical results for the Long Term Average CD . . . 56
2.7.2 Numerical Results for the Effective Data Rate . . . 58
2.8 Conclusions . . . 61
References . . . 62
3 Application of Change Point Analysis in Edge Resource Allocation and Intrusion Detection 68 3.1 Introduction . . . 68
3.2 Contributions and Chapter Organization . . . 69
3.2.1 CP Analysis in Resource Allocation . . . 69
3.2.2 CP Analysis for Anomaly Detection in SDWSNs . . . 71
3.2.3 Chapter Organization . . . 71
3.3 Related Works . . . 71
3.4 Training (Off-line) Phase . . . 73
3.4.1 Basic Off-line Approach . . . 73
3.4.2 Extended Off-line Approach . . . 75
3.5 On-line Phase . . . 75
3.5.1 On-line Analysis . . . 75
3.5.2 Trend Indicator . . . 77
3.5.3 Overall Algorithm . . . 78
3.6 Validation of the RCPD Using Synthetic Data . . . 79
3.7 Performance Evaluation Using Real Data . . . 83
3.7.1 Statistical Properties of the Real Dataset . . . 83
3.7.2 Performance of the Off-line Training Phase . . . 84
3.7.3 Evaluation of the RCPD Algorithm . . . 85
3.7.4 Time Dependencies of Piecewise time-series . . . 87
3.7.5 Computational Complexity and Scalability . . . 88
3.8 The RCPD Algorithm in a Load Balancing Scenario . . . 89
3.9 Application of the RCPD for Intrusion Detection in SDWSNs . . . 90
3.9.1 SDWSN Security Analysis . . . 91
3.9.2 Impact of DDoS Attacks on Network Performance . . . 91
3.9.3 RCPD for Intrusion Detection . . . 92
3.10 Results and Analysis . . . 93
3.10.1 FDFF Attack Detection . . . 93
3.10.2 FNI Attack Detection . . . 98
3.11 Conclusion . . . 99
References . . . 100
4 Uplink Non-Orthogonal Multiple Access (NOMA) Under Statistical QoS Delay Constraints 104 4.1 Introduction . . . 104
4.2 Contributions and Chapter Organization . . . 104
4.3 Effective Capacity of Two-user NOMA Uplink Network . . . 105
4.3.1 ECs in a Two-user NOMA Uplink Network . . . 106
5 Perspectives 119
5.1 Introduction . . . 119
5.2 The Role of of PLS in 6G . . . 120
5.2.1 How Many Secret Bits Are Needed . . . 120
5.2.2 Authentication . . . 121
5.2.3 Data Confidentiality . . . 121
5.2.4 Anomaly Detection . . . 122
5.3 Low Latency, Interference-free, Contextual 6G Communications . . . 122
5.3.1 NOMA for Collision Avoidance in mMTC Uplink . . . 122
5.3.2 Interference Cancellation Using Machine Learning . . . 123
5.3.3 Towards Context Aware Communications in 6G . . . 123
References . . . 124
3.1 Percentage of the successful CP detections for the standard and modified BS algorithm 80
3.2 Success rates of trend indicators . . . 80
3.3 Results of the RCPDs’ algorithm CPs detection for one change in the mean value. . . 81
3.4 Results of the RCPDs’ algorithm CPs detection for two mean changes. . . 82
3.5 Success rates of T If trend indicator . . . 84
3.6 Empirical percentiles of mean values change rate. . . 87
3.7 Percentages of time-series with Time Dependencies Exceeding t Samples . . . 89
3.8 Simulation Parameters . . . 92
3.9 FDFF Attack Detection, 36 Nodes, 5% Attackers . . . 94
3.10 FDFF Attack Detection, 100 nodes, 5% Attackers . . . 94
3.11 FDFF Attack Detection, 36 nodes, 20% Attackers . . . 95
3.12 FDFF Attack Detection, 100 nodes, 20% Attackers . . . 95
3.13 FNI Attack Detection, 36 nodes, 5% Attackers . . . 96
3.14 FNI Attack Detection, 100 nodes, 5% Attackers . . . 96
3.15 FNI Attack Detection, 36 nodes, 20% Attackers . . . 97
1.1 Recent research areas and topics . . . 16
2.1 Roadmap of contributions. . . 41
2.2 Secret key generation between Alice and Bob. . . 44
2.3 Pipelined SKG and encrypted data transfer between Alice and Bob. . . 46
2.4 a) Efficiency comparison for N = 12, SNR=10 dB and κ = 2. . . 56
2.4 b) Efficiency comparison for N = 64, SNR=10 dB and κ = 2. . . 56
2.5 Efficiency vs κ, for N = 24, SNR=10 dB. . . 57
2.6 a) Size of set D for different SNR levels and σ2 e when N = 24. . . 57
2.6 b) Size of setD for different values of κ when N = 24. . . . 57
2.7 a) Effective data rate achieved by the parallel heuristic approach and the sequential approach when N = 12, SNR= 10 dB and κ = 2. . . 58
2.7 b) Effective data rate achieved by the parallel heuristic approach and the sequential approach when N = 64, SNR= 10 dB and κ = 2. . . 58
2.8 a) Effective data rate achieved by the parallel heuristic approach and the sequential approach when N = 12, SNR= 0.2 dB and κ = 2. . . 59
2.8 b) Effective data rate achieved by the parallel heuristic approach and the sequential approach when N = 64, SNR= 0.2 dB and κ = 2. . . 59
2.9 a) Effective data rate achieved by parallel and sequential approaches when N = 12, SNR= 5dB, θ = 0.0001, κ = 2. . . 60
2.9 b) Effective data rate achieved by parallel and sequential approaches when N = 12, SNR= 5dB, θ = 100, κ = 2. . . 60
2.9 c) Effective data rate achieved by parallel and sequential approaches when N = 64, SNR= 5dB, θ = 0.0001, κ = 2. . . 60
2.9 d) Effective data rate achieved by parallel and sequential approaches when N = 64, SNR= 5dB, θ = 100, κ = 2. . . 60
3.1 Estimated a) frequency and b) cumulative frequency of the number of CPs per time-series. 84 3.2 Frequency values of the number of upward and downward CPs, per time-series. . . 85
3.3 a) Boxplot including the interval (5%− 95%) (dashed line) and (10% − 90%) interval (dotted line), b) Cumulative frequency for the interim time of consecutive CPs. . . 86
3.4 DTW distances for the two on-line detection schemes. . . 86
3.5 Outputs of the RCPD algorithm using standard CUSUM for different time-series. Solid and dashed lines depict an upward and a downward change, respectively. . . 87
3.6 Outputs of the RCPD algorithm using standard type CUSUM for different time-series. Solid and dashed lines depict an upward and a downward change, respectively. . . 88
3.7 Outputs of the RCPD algorithm using ratio type CUSUM for different time-series. Solid and dashed lines depict an upward and a downward change, respectively. . . 88
3.8 Outputs of the RCPD algorithm; using ratio type CUSUM for different time-series. Solid and dashed lines depict an upward and a downward change, respectively. . . 89
3.10 a) time-series of video content views, red lines depict the detected CPs, b) the connection time with and without RCPD adaptation and c) the equivalent servers’ CPU utilization. 90 4.1 E1
c, Ec2in a two-user NOMA uplink network compared to Ecs of two users OMA, versus ρ109
4.2 Ec1 versus the transmit SNR, for several delays. . . 109
4.3 Ec2 versus the transmit SNR ρ for several delays. . . 110
4.4 E1 c and Ec2 in a two-user NOMA compared to ECs of two users OMA, versus normalized delay β, for different values of ρ. . . 110
4.5 Ec1− ˜Ec1 versus ρ, for several values of the normalized delay exponent. . . 111
4.6 E2 c − ˜Ec2 versus ρ, for various normalized delay exponent. . . 111
4.7 VN and VO versus ρ, for several values of normalized delay exponent. . . 112
4.8 VN - VO versus ρ for various normalized delay. . . 113
List of Abbreviations 0-RTT Zero round trip time
3GPP The 3rd Generation Partnership Project 5G Fifth generation
6G Sixth generation
AE Authenticated encryption
AES GCM Advanced encryption standard Galois counter mode ARMA Autoregressive moving average model
B5G Beyond 5G
BF-AWGN Block fading additive white Gaussian noise CRP Challenge response pair
CSI Channel state information DMT Detection median time DR Detection rate
EAP-TLS Extensible authentication protocol - transport layer security EC Effective capacity
EH Energy harvesting
FDFF False data flow forwarding FNI False neighbour information FNR False negative rate
FPR False positive rate FTN Faster than Nyqist
GARCH Generalized autoregressive conditional heteroskedasticity HMAC Hashed message authentication code
IoT Internet of things LRT Likelihood ratio test MA Multiple access MAC Media access control MiM Man in the middle
mMTC massive machine type communications NB-IoT Narrow band IoT
NOMA Non Orthogonal multiple access
OFDM Orthogonal frequency division multiplexing OMA Orthogonal multiple access
PAM Pulse amplitude modulation PHY Physical layer
PKE Public key encryption PLS Physical layer security
PNC Physical layer network coding PUF Physical unclonable function QAM Quadrature amplitude modulation QoS Quality of service
QoSec Quality of security RAN Radio access network
RPL Routing protocol for low-power and lossy networks RSA Rivest, Shamir, Adleman
RSS Received signal strength SC Secrecy capacity
SDN Software defined networking
SDWSN Software defined wireless sensor network SIR Signal to interference ratio
SKG Secret key generation SLA Service level agreement SNR Signal-to-noise ratio
URLLC Ultra reliable low latency communication V2X Vehicle-to-everything communication WSN Wireless sensor network
Activity Review
1.1
Motivation for Application for the HdR Diploma
With this thesis, I wish to submit my application for the Habilitation to Direct Research at CY -Cergy Paris Universit´e. Currently, I am a Maˆıtre de Conf´erences at the Ecole Nationale Sup´erieure de l’Electronique et de ses Applications (ENSEA) in Cergy and in parallel I have a Visiting Research Fellow status at the Department of Electrical and Electronic Engineering of Princeton University in the USA and at the School of Computer Science and Electronic Engineering of the University of Essex in the UK.
Wireless
Communications
5G, B5G
Security
Multi-factor authentication
Signal Processing
Stochastic SP - PLS - Secret key generation - PUFs - Resource allocation - NOMA - - Low latency Intrusion detection MiM - Proximity estimation - Jamming - Injection attacksFigure 1.1: Recent research areas and topics
My current research activities relate to various topics in wireless communications and physical layer security with an emphasis on the proposal of low latency communication systems and the development of new security protocols for future generations of wireless. I actively work on topics in flexible numerology, non-orthogonal mutiple access (NOMA) and fast authentication protocols for
delay constrained systems using physical unclonable functions (PUFs) and RF fingerprinting. In this framework, with my current research team, that comprises four PhD students and two postdoctoral researchers, we investigate resource allocation in beyond fifth generation (B5G) leveraging NOMA, the efficient design of Slepian Wolf and Wyner Ziv reconciliation decoders at the short block-length, the development of zero-round-trip-time (0-RTT) authentication protocols using resumption keys generated from wireless fading coefficients, the analysis of the wireless channel secrecy capacity under statistical delay quality of service (QoS) constraints and the development of quick anomaly detection algorithms for software defined networks.
My research lies at the interface of wireless communications, signal processing and security studies, as depicted in Fig. 1.1; at this – not so-frequented – scientific crossroad, new engineering problems are encountered, a few of which will be discussed in later chapters of this thesis, along with proposed solutions.
With respect to my contribution as an academic teacher and supervisor, I have a long experience in teaching and supervising students in security, coding and wireless communications for more than 8 years in the UK and France. I have had the chance to teach a variety of courses both at the undergraduate and graduate level and contribute in teaching in the continuous education engineering track of ENSEA. I have taught to a variety of class sizes and have customarily received very positive feedback from my students, both in formal assessment and in face-to-face interaction. Furthermore, since last September I am acting as the liaison of the international mobility for ENSEA students towards the UK and have secured internships at Imperial College London, the University of York, etc.
In my academic employment I have had the opportunity to undertake a number of important administrative responsibilities. I currently head the research team ICI (information, communications, imaging) of the ETIS Lab that comprises 13 permanent faculty (3 PU, 8 MCF, 2 CNRS CR) and more than 18 research students and teaching fellows. In this role my aim is to help maintain and enhance the quality and quantity of the team’s collective research output, its ability to attract research funding and good young researchers, increase further the team’s visibility in the national and international level and ensure the team members work in a friendly and fertile learning environment. Furthermore, during my employment at the School of Computer Science and Electronic Engineering in the UK, I have acted in 2017 as the President of the Athena Swan Committee, steering 15 faculty and admin staff for the preparation of the department’s gender equality and diversity charter.
With respect to my involvement in professional bodies, I am a member of the IEEE INGR Roadmap Security Workgroup, of the IEEE P1940 Standardization Workgroup on ”Standard profiles for ISO 8583 authentication services” and have been a member of the IEEE Teaching Awards Committee for the last three years.
I feel that my overall experience is of sufficient standing to allow me to lead independent research and act as a stand-alone thesis advisor / director. A brief overview of my research and teaching activities the last 8 years is provided in Sections 1.4 and 1.5 respectively. First, I introduce myself to the reader through a detailed academic CV in Section 1.2. and the full list of my publications in Section 1.3.
1.2
Curriculum Vitae
In the following pages my full academic Curriculum Vitae is provided, including a full record of my publications in Section 1.3. The interested reader may also consultmy web pageandmy google scholar page.
Address: Room 341, ENSEA, 6 Avenue du Ponceau, Cergy, FR Telephone: +33 (0)769113367
e-mail: arsenia.chorti@ensea.fr achorti@princeton.edu
1.2.1 Current Position / Responsibilities (in chronological order)
Sep. 2017-present: ENSEA (ETIS UMR8051) Associate Professor (MCF) in Communications and Networking, Research Group: 4 PhD students, 2 postdocs
Sep. 2017-Jul. 2020: Member of the IEEE Teaching Awards Committee
Sep. 2019 – present: PEDR (prime d’encadrement doctoral et recherche) – premium for excellence in supervision and research
Sep. 2019 – present: Member of the IEEE P1940 Standardization Workgroup on "Standard profiles for ISO 8583 authentication services"
April 2020-present: Head of the Information, Communications and Imaging (ICI) Group of ETIS UMR 8051 (Responsable d’équipe information, communications et imagerie), comprising 3 Professors, 8 MCF, 2 CNRS Researchers, 2 Postdocs, 2 ATERs, 14 PhD students
Apr. 2020: Award of CNRS delegation (half year travel sabbatical) to visit Prof. H.V. Poor (Princeton University, NJ USA) and Dr. A. Barolo (Barkhausen Institute, Dresden DE) in Spring / Summer 2021
May 2020: Member of the IEEE International Network Generations Roadmap (INGR) Security Workgroup (pre-standardization workgroup for security in future networks)
June 2020: Elevated to IEEE Senior Member
1.2.2 Research Interests
My current research spans the areas of wireless security and beyond fifth generation (B5G) networks. I work on the design of security schemes for B5G, with a particular focus on physical layer security; my recent contributions concern fast authentication protocols using physical unclonable functions (PUFs) and secret key generation (SKG) from shared randomness, with proximity / localization as an extra authentication factor. Furthermore, I work on low latency communications, leveraging recent results on non-orthogonal multiple access (NOMA), investigate polynomial complexity algorithms for flexible numerology and eMBB – URLLC coexistence and joint PHY-MAC resource allocation optimization using the theory of the effective capacity. Recent contributions (since 2017) include:
o Wireless security for B5G and Internet of things (IoT) [J19], [J21], [C37], [C33], [S2] o Authentication protocols leveraging PUFs, SKG and proximity estimation [BC3], [U1] o Resource allocation in 5G using change point analysis [J17], [J20], [C32]
o Anomaly detection in software defined networks [C39], [J18], [S1], [U2] o Active attacks in PHY [J15], [J16], [C36], [C28-C31] [BC2]
o Low latency B5G communications, non-orthogonal multiple access (NOMA), NOMA-R, flexible numerology for B5G [J22], [U3-U5]
1.2.3 Education
2000-2005 Imperial College London: Department of Electrical and Electronic Engineering Ph.D. in Communications and Signal Processing
Thesis Title: “The Impact of Circuit Nonlinearities and Noise in OFDM Receivers”, Supervisor: Mike Brookes, Scholarship awarded by Ι.Κ.Υ. and Panasonic UK ltd.
1999-2000 Université Pierre et Marie Curie – Paris VI MSc (D.E.A.) in Electronics
Dissertation Title: “F.P.G.A. Implementation of Multi-Layer Perceptron Neural Network for Real-Time Applications in High Energy Physics”, Supervisor: Prof. Patrick Garda, Ι.Κ.Υ. Scholar
an Industrial Network”, Supervisor: Prof. K. Koumbias
1.2.4 Academic Employment
o September 2017 – present: ENSEA, Associate Professor in Communications and Networks o October 2013 – August 2017: University of Essex, School of Computer Science and Electrical
Engineering, Lecturer in Communications and Networks and subsequently Visiting Research Fellow (ongoing)
o November 2012 – October 2013: Foundation for Research and Technology Hellas (FORTH), Institute of Computer Science, International Outgoing Fellow (IOF) Marie Curie Research Fellow
o May 2011 – present: Princeton University, Dep. of Electrical Engineering, IOF Marie Curie Fellow and subsequently Visiting Research Fellow
o December 2008 – April 2011: Middlesex University UK, School of Engineering and Information Sciences, Dep. of Computer Communications, Senior Lecturer in Communications and Networks
o October 2007 – September 2009: University College London (UCL), Dep. of Electronic and Electrical Engineering, Postdoctoral Research Fellow and subsequently Visiting Researcher o October 2006 – September 2007: Technical University of Crete (TUC), Department of Mineral
Resources Engineering, Resources Detection and Identification Research Unit, Postdoctoral Research Fellow
o October 2005 – September 2006: University of Southampton, School of Electronics and Computer Science, Electronic Systems Design Group (ECS), Postdoctoral Research Fellow
1.2.5 Research Funding and Grants
Project proposals currently under review:
o Principal Investigator (PI) of ANR PRCE project HERCULES (enhancement measures in the security of beyond fifth generation networks) 2nd round AAPG 2020, with the SME Montimage, K. Salamatian (LISTIC), I. Andriyanova (ETIS), A. Histace (ETIS) and F. Ghaffari (ETIS)
o External collaborator of project LEON (Intelligent Network Softwarization for the Internet of Things), ELIDEK, GR, with Dr. L. Mamatas
o PI project PROCOPE PHC (travel grant) to visit the Barkhausen Institute, DE in 2021-2022 Ongoing projects:
o Co-investigator (co-I) project PHEBE (Physical layer security for beyond fifth generation communications) with L. Wang (PI), L. Luzzi, M. Chafii, M. Le Treust, Paris-Seine Excellence Initiative, 2020-2024, 400,000€
o PI project SAFEST with F. Jardel (NOKIA Bell Labs) (Physical layer security for future generations wireless systems), DIM RFSI, 2019-2021, 27,500€
o PI project eNiGMA (Non-orthogonal multiple access techniques under security and delay constraints), with I. Fijalkow, Paris-Seine Excellence Initiative, 2019-2021, 110,000€
o Co-I project ELIOT (Enabling technologies for IoT), ANR PRCI with Univ. Sao Paolo, Brazil, with V. Belmega (PI), I. Andriyanova, I. Fijalkow, J. Lorandel, Role: Leader of WP on IoT security, 2019-2023, ETIS: 390,420€ (total of 740 k€)
Past projects:
o PI SRV-ENSEA de l’Institut des Etudes Avancées Université Paris Seine, 2018-2019: 3,000€ o PI SRV-ENSEA Institut des Etudes Avancées Université Paris Seine: 2017-2018 : 2,850€ o PI project PHOTINO, University of Essex, Research and Innovation Fund: 2014-2015: £13,000
o PG Scholarship from the State Scholarships Foundation of Greece–I.K.Y. 2000-2004: £41,820
1.2.6 Teaching and Related Responsibilities at ENSEA (since 2017)
o 2019-present: Responsible of student international mobility to the UK, 2nd year MEng, 3rd
year MEng, Erasmus programme with the UK
o 2019-present Instructor in the MSc (M2R) module “Cryptography and Network Security”, University Cergy Pontoise, Master 2 Informatique et Ingénierie des Systèmes Complexes (IISC), specilization SIC (Signal, Information, Communications),
o 2018-present: Responsible of the module “Network security” 3rd year MEng, ENSEA
o 2018-present: Responsible of module “Interconnexion réseaux” 3rd year Cycle par
Alternance, ENSEA
o 2017-present: Responsible of the Option Internet of Things “Option IoT”, 2nd year MEng,
ENSEA
o 2017-present: Instructor “IoT Security”, 2nd year MEng, ENSEA
o 2017-present: Responsible of the module “Internetworking”, 3rd year MEng, ENSEA
o 2017-present: Instructor “Wireless Communications”, 3rd year MEng, ENSEA
o 2017-present: Lab instructor in various courses, including Digital Communications, Internetworking, Signals and Systems, etc.
1.2.7 Research Supervision
Current supervision
o PhD Student Mr. Miroslav Mitev: supervision @60%, 25/4/2017-9/2020, "Physical layer security for the Internet of things", co-supervised with Dr. M. Reed, University of Essex, UK, Thesis VIVA (defence) scheduled for September 2020, publications: [J21], [C37], [C36], [C33], [P1], [U1]
o PhD student Mr. Sotiris Skaperas: supervision @40%, 1/9/2017-9/2020, "Data analysis and forecasting models for flexible resource management in 5th generation networks", co-supervised with Dr. L. Mamatas, University of Macedonia in Thessaloniki, GR, Thesis defence scheduled for September 2020, publications: [J20], [J17], [C32], [U5]
o PhD student Mr. Gustavo Alonso Nunez Segura: supervision @35%, 1/2/2019-projected to finish in 1/2022 (4-year thesis programme in Brazil), "Cooperative Intrusion Detection System for Software Defined Wireless Sensor Networks", co-supervised with Prof. Cintia Borges Margi, University of Sao Paolo, Brazil, publications: [J18], [C39], [C35], [S1], [U2]
o PhD student Mr. Mouktar Bello: supervision @70%, 1/11/2020-projected to finish in 10/2023, "Meeting delay and security constraints in 6G wireless networks", co-supervised with Prof. I. Fijalkow, ETIS/ENSEA, FR, publications: [C38], [U3, U4]
o Postdoc Dr. Mahdi Shakiba Herfeh: supervision @100%, 21/11/2019-20/5/2021 (fixed term 1.5 years), “Physical layer security for IoT applications”, project ELIOT ANR PRCI, ETIS/ENSEA FR, publications: [BC3], [U1]
o Postdoc Dr. Nasim Ferdosian: supervision @90%, 1/1/2020-31/12/2021 (fixed term 2 years), “Non-orthogonal multiple access techniques under security and delay constraints”, with Prof. I. Fijalkow, ETIS/ENSEA, FR, publications: [U5]
Past supervision
o MSD (Master by Thesis – full year research project) student Cornelius Saiki: supervision @84%, 1/9/2014-31/8/2015, “A Novel Physical Layer Key Generation and Authenticated Encryption Protocol Exploiting Shared Randomness”, co-supervised with Prof. S. Walker, University of Essex, publications: [C27]
o MSc (M2R) SIC student Rihem Nasfi: supervision @100%, 1/11/2018-15/3/2019, Projet d’Initiation à la Recherche (PIR), “Non-orthogonal multiple access networks under QoS delay constraints”, publications : [C34]
o MSc (M2R) SIC student Gada Rezgui, supervision @50%, 1/11/2016-15/3/2017, Projet d’Initiation à la Recherche (PIR), “Secret Key Generation systems under Jamming Attacks via Game Theoretic Tools”
o MSc (M2R) IMD student Amani Gran, supervision @100%, 1/11/2018-15/3/2019, Projet d’Initiation à la Recherche (PIR), “IoT lightweight security”
o MSc (M2R) SIC student Fatiha Ait Larbi, supervision @100%, 1/11/2018-15/3/2019, Projet d’Initiation à la Recherche (PIR), “Cross-layer security protocol design”
o MSc (M2R) SIC student Mouad Nahri, supervision @100%, 1/11/2019-15/3/2020, Projet d’Initiation à la Recherche (PIR), “Flexible numerology for B5G”
o Other MSc/BSc supervision: 5 MSc and 9 BSc dissertations at the University of Essex and more than 10 MSc and BSc dissertations at Middlesex University
1.2.8 Recruitment (Selection) Committees / Thesis Examiner
o May 2020: Recruitment Committee (Comité de sélection) for a MCF post at CY Cergy University on Networks and Security
o Sep. 2019: Recruitment Committee (Comité de sélection) for a MCF post at EISTI on Cybersecurity
o Jun. 2020: Thesis Examiner (rapporteur), A. Ben Hadj Fredj, Télécom ParisTech, supervisors Prof. G. Rekaya and Prof. J-C Belfiore, “Computations for Multiple Access Channels in Wireless Networks”
o Jan. 2019: Thesis Reviewer, L. Senigagliesi, Univ. Polytechnica delle Marche, supervisors Prof. L. Spalazzi and Prof. M. Baldi, “Information-theoretic security techniques for data communications and storage”
o Aug. 2014: Thesis Examiner, I. K. Musa, CSEE University of Essex UK, supervisor Prof. S. Walker, “Optimized Self-Service Resource Containers for Next Generation Cloud Delivery”
1.2.9 Workshop Organization / Keynotes / Tutorials
o Tutorial on “What Physical Layer Security Can Do for 6G”, IEEE Global Communications (GLOBECOM) 2020, A. Chorti and V.H. Poor, December 2020, Taipei TW.
o Tutorial on “Statistical methods in physical layer security”, IEEE Statistical Signal Processing (SSP) Workshop, July 2020, Rio de Janeiro, BR (rescheduled to July 2021 due to COVID-19) o Special Session Organizer, “Selected topics on 6G security”, IEEE ISWCS, Sep. 2020, Berlin,
Germany (rescheduled to Sep. 2021 due to COVID-19)
o Special Session Organizer, “Statistical Methods for IoT”, IEEE SSP 2020, Jul. 2020, Rio de Janeiro, Brazil (postponed to July 2021 due to COVID-19)
o Training School Co-organizer (with M. Chafii, S. Stanczak and R. Cavalcante), “Machine Learning for Communications”, 3-4 Sep. 2020, Berlin (co-located with ISWCS, rescheduled to Sep. 2021 due to Covid-19)
o Chair of the GdR ISIS Workshop “Women in Communications, Information Theory and Signal Processing”, May 19 2020 (rescheduled to May 2021 due to Covid-19)
o Chair of the GdR ISIS Workshop “Enabling ultra-reliability, low latency and massive connectivity”, June 18 2020 (virtual event due to Covid-19)
o Keynote IEEE PIMRC Workshop Security Public RATs: “Practical examples of physical layer security”, 4 Sep. 2016, Valencia, Spain
o Co-chair of “2nd Women’s Workshop on Communications and Signal Processing”, 16-18 July 2014, Princeton NJ, US
o Track chair of the IEEE Global Wireless Summit 2014, 11-14 May 2014, Aaborg, Denmark o Chair of the “Second International Conference on Communications, Connectivity,
Convergence, Content and Cooperation”, 11-14 May 2014, Aalborg, Denmark o Chair of the “WirelessVITAE, 10-13 May 2014, Aalborg, Denmark
1.2.10 Editor / Reviewer / Selected TPCs
o 2020- present: Associate Editor of the IEEE Open Journal on Signal Processing (OJSP)
o Sep. 2019-present: Lead Guest Editor, EURASIP JWCN Special Issue “Physical layer security solutions for 5G-and-beyond”, Editors: S. Tomasin, H.V. Poor, M. Baldi, S. El Ruayheb, X. Wang, to appear in 2020
o 2018-2019: Executive Editor Transactions on Emerging Telecommunications Technologies (ETT), Wiley
o 2017-2019: Executive Editor of Internet Technology Letters (ITL), Wiley
o Reviewer: IEEE Transactions (Trans.) on Information (Inf.) Forensics and Security, Elsevier Computers and Security, IEEE Trans. on Wireless Communications (Commun.)., IEEE Trans. Signal Processing, IEEE Trans. Vehicular Technologies, IEEE JSAC, IEEE Wireless Commun. Letters (L.), IEEE Commun. L., Trans. on Emerging Telecom Tech. (ETT), Eurasip JWCN, IEEE Trans on Commun., …
o TPCs: more than 30 TPCs, indicatively IEEE GLOBECOM 2015, 2016, 2017, 2018, 2019, 2020, IEEE ICC 2014, 2015, 2016, 2018, 2019, 2020, IEEE WCNC 2016, 2019 (executive member), …
1.2.11 Selected Invited Talks (after 2016)
o July 2019, “Physical layer security in delay constrained applications”, NOKIA Bell Labs, FR o May 2019, “Physical layer security in delay constrained applications”, Barkhausen Institute,
Dresden DE
o May 2019, “Physical layer security in delay constrained applications”, ICS FORTH, GR o October 2017, “Emerging security paradigms”, Thales, FR
o March 2017, “Physical layer security for future networks”, British Telecom, Adastral Park, UK o June 2016, “Practical examples of physical layer security”, Summer Research Institute, EPFL,
CH
1.2.12 Past Administrative Responsibilities and Outreach Activities
o 2016-2017: President of the Committee for Gender Equality and Diversity Athena Swan, Univ. Essex, UK
o 2016-2017: Vice-president “Research Student Progress and Management Committee”, Univ. Essex, UK
o 2015-present: Fellow of the Higher Education Academy, UK (professional title in pedagogical training)
1.3
Publication List
1.3.1 Books [B] / Book Chapters [BC]
(supervised students and postdocs appear underlined)
BC3 M. Shakiba Herfeh, A. Chorti, V.H. Poor, A Review of Recent Results on Physical Layer Security, to appear in Springer Nature 2020;
BC2 A. Chorti, A Study of Injection and Jamming Attacks in Wireless Secret Sharing Systems, (Proc. 2nd Workshop Communication Security, WCS 2017), Lect. Notes in Elect. Eng., vol 447,
pp. 1-14, Springer;
BC1 A. Chorti, C. Hollanti, J.-C. Belfiore, H.V. Poor, Physical Layer Security: A Paradigm Shift in Data Confidentiality, Springer, Lecture Notes in Electrical Engineering - Physical and Data-Link Security Techniques for Future Communication Systems, vol. 358, pp. 1-15, Sep. 2015;
B A. Chorti, The Impact of Circuit Nonlinearities and Noise in OFDM Receivers, Feb. 2010, Verlag
1.3.2 Refereed International Journals [J]
(supervised students and postdocs appear underlined)
J22 M. Pischella, A. Chorti, I. Fijalkow, ”On the Performance of NOMA-Relevant Strategies Under Statistical Delay QoS Constraints”, IEEE Wireless Commun. Letters, in print (early access); J21 M. Miroslav, A. Chorti, M.J. Reed, L. Musavian, ”Authenticated Secret Key Generation in
Delay Constrained Wireless Systems”, EURASIP J Wireless Com Network, vol. 122, Jun. 2020; J20 S. Skaperas, L. Mamatas, A. Chorti, ”Real-Time Algorithms for the Detection of Changes in the Variance of Video Content Popularity”, IEEE Access, vol. 8, pp: 30,445-30,457, Feb. 2020; J19 W. Yu, A. Chorti, L. Musavian, V.H. Poor, Q. Ni, ”Effective Secrecy Capacity for a Downlink NOMA Network”, IEEE Trans. Wireless Commun., vol. 18, no 12, pp: 5,673-5690, Dec. 2019; J18 G.A. Nunez Segura, C. B. Margi, A. Chorti, ”Understanding the Performance of Software Defined Wireless Sensor Networks Under Denial of Service Attack”, Open Journal of Internet of things (OJIOT), Vol.5, no 1, pp:59-68 Aug. 2019 (published in the OJIOT as a special issue); J17 S. Skaperas, L. Mamatas, A. Chorti, ”Real-Time Video Content Popularity Detection Based
on Mean Change Point Analysis”, IEEE Access, vol.7 pp: 142,246-142,260, Jul. 2019;
J16 G. Rezgui, E.V. Belmega, A. Chorti, ”Mitigating Jamming Attacks Using Energy Harvesting”, IEEE Wireless Commun. Let., vol. 8 no 1, pp: 297-300, Feb. 2019;
J15 E.V. Belmega, A. Chorti ”Protecting Secret Key Generation Systems against Jamming: Energy Harvesting and Channel Hopping Approaches”, IEEE Trans. Inf. Forensics Security, vol. 12, no 11, pp: 2611-2626, Nov. 2017;
J14 D. Karpuk, A. Chorti, ”Perfect Secrecy in Physical-Layer Network Coding Systems from Structured Interference”, IEEE Trans. Inf. Forensics Security, vol. 11, no 8, pp. 1875-1887, Aug. 2016;
J13 A. Chorti, K. Papadaki, H.V. Poor, ”Optimal power allocation in block fading channels with confidential messages”, IEEE Trans. Wireless Commun., vol. 14, no 9, pp. 4708-4719, Sep. 2015;
J12 A. Chorti, S. Perlaza, Z. Han, H.V. Poor, ”On the resilience of wireless multiuser networks to passive and active eavesdroppers”, IEEE Journal of Selected Areas in Commun., vol. 31 no 9, pp. 1850-1863, Sep. 2013;
J11 A. Chorti, M. Brookes, ”On the effect of Voigt profile oscillators on OFDM systems”, IEEE Trans. Circuits Syst. II, vol. 58, no 11, pp. 768-772, Nov. 2011;
J10 G. Spiliopoulos, D.T. Hristopulos, M.P. Petrakis, A. Chorti, ”A multigrid method for the estimation of geometric anisotropy in environmental data from sensor networks”, Elsevier Computers and Geosciences, vol. 37, no 3, pp. 320-330, Mar. 2011;
J9 A. Chorti, M. Brookes, ”Performance Analysis of COFDM and DAB Receivers in narrow-band and tonal interference”, Springer Telecommunication Systems J., vol. 46, no 2, pp. 181-190, 2011. J8 Y. Kanaras, A. Chorti, M. Rodrigues, I. Darwazeh, ”A fast constrained sphere decoder for ill
conditioned communication systems”, IEEE Commun. Let., vol. 14, no 11, pp. 999-1001, Nov. 2010.
J7 A. Chorti, ”How to model the near-to-the-carrier regime and the lower knee frequency of real RF oscillators”, J. Electrical Computer Eng., vol. 2010, article ID 537132, Oct. 2010.
J6 A. Chorti, D.T. Hristopulos, ”Non-parametric identification of anisotropic correlations in spatially distributed data sets”, IEEE Trans. Signal Proces, vol. 56, no 10, pp. 4738-4751, Oct. 2008.
J5 D. Karantzas, A. Chorti, N.M. White, C.J. Harris, ”Teaching old sensors new tricks: archetypes of intelligence”, IEEE Sensors J., Special Issue on Intelligent Sensing”, invited paper, vol. 7, no 5, pp. 868-881, May 2007.
J4 A. Chorti, D. Karatzas, N.M. White and C.J. Harris, ”Intelligent Sensors in Software: The Use of Parametric Models for Phase Noise Analysis”, International Journal of Information Processing, vol. 1, no. 2, June 2007.
J3 A. Chorti, D. Karantzas, N.M. White and C.J. Harris, ”Use of the extended Kalman filter for state dependent drift estimation in weakly nonlinear sensors”, Sensors Let., vol. 4, no 4, pp. 377-379, Dec. 2006.
J2 A. Chorti, M. Brookes, ”A spectral model for RF oscillators with power-law phase noise, IEEE Trans. Circuits Syst. I ”, vol. 53, no 9, pp. 1989-1999, Sep. 2006.
J1 A. Chorti, M. Brookes, ”On the effects of memoryless nonlinearities on M-QAM and DQPSK OFDM Signals”, IEEE Trans. Microw. Theory Techn., vol. 54, no 8, pp. 3301-3315, Aug. 2006.
1.3.3 Refereed International Conference Proceedings [C]
(supervised students and postdocs appear underlined)
C39 G.A. Nunez Segura, S. Skaperas, A. Chorti, L. Mamatas, C. Borges Magri, “Denial of Service Attacks Detection in Software-Defined Wireless Sensor Networks”, Proc. IEEE Int. Conf. Commun. (ICC) Worskhop on SDN Security, Dublin UK, 7-11 Jun. 2020;
C38 B. Mouktar, W. Yu, A. Chorti, L. Musavian, “Performance Analysis of NOMA Uplink Networks under Statistical QoS Delay Constraints”, Proc. IEEE Int. Conf. Commun. (ICC), Dublin UK, 7-11 Jun. 2020;
C37 M. Mitev, A. Chorti, M.J. Reed “Subcarrier Scheduling for Joint Data Transfer and Key Genera-tion Schemes in Multicarrier Systems”, Proc. IEEE Int. Global Commun. Conf. (GLOBECOM), Hawaii US, 9-13 Dec. 2019;
C36 M. Mitev, A. Chorti, E.V. Belmega, M.J. Reed “Man-in-the-Middle and Denial of Service Attacks in Wireless Secret Key Generation”, Proc. IEEE Global Commun. (GLOBECOM), Hawaii US, 9-13 Dec. 2019;
C35 G.A. Nunez Segura, C. B. Margi, A. Chorti , “Understanding the Performance of Software Defined Wireless Sensor Networks Under Denial of Service Attack”, Proc. Int. Workshop on Very Large IoT (VLIoT) 2019, Los Angeles, US, 30th Aug. 2019 (*invited paper);
C34 R. Nasfi, A. Chorti, “Performance Analysis of the Uplink of a Two User NOMA Network under QoS Delay Constraints”, Proc. IEEE Int. Conf. on Ubiquous and Future Networks (ICUFN) 2018, Zagreb, Croatia, 2-5 July 2019;
C33 M. Mitev, A. Chorti, M.J. Reed “Optimal Resource Allocation in Joint Secret Key Generation and Data Transfer Schemes”, Proc. IEEE Int. Conf. Wireless Commun. Mobile Comput. (IWCMC), Tangiers Morocco, 24-28 June 2019;
C32 S. Skaperas, L. Mamatas, A. Chorti, “Early Video Content Popularity Detection with Change Point Analysis”, Proc. IEEE Int. Global Commun. (GLOBECOM), Abu Dhabi, UAE, 6-11 December 2018;
C31 E.V. Belmega, A. Chorti, “Energy Harvesting in Secret Key Generation Systems under Jamming Attacks”, Proc. IEEE Int. Conf. Commun. (ICC), Paris, France, May 2017;
C30 A. Chorti, “Secret Key Generation in Rayleigh Block Fading AWGN Channels under Jamming Attacks”, Proc. IEEE Int. Conf. Commun. (ICC), Paris France, May 2017;
C29 A. Chorti, “Optimal Signalling Strategies and Power Allocation for Wireless Secret Key Gener-ation Systems in the Presence of a Jammer”, Proc. IEEE Int. Conf. Commun. (ICC), Paris, France, May 2017;
C28 A. Chorti, “Overcoming limitations of secret key generation in block fading channels under active attacks”, Proc. IEEE 17th Int. Workshop Signal Process. Advances Wireless Commun. (SPAWC), pp. 1-5, Jul. 2016 (*invited paper);
C27 C. Saiki, A. Chorti, “A novel authenticated encryption protocol exploiting shared randomness”, Proc. IEEE Commun. Network Security (CNS), 2nd Workshop on Physical Layer methods for Wireless Security, pp. 651-656, Sep. 2015;
C26 A. Chorti, M.M. Molu, D. Karpuk, C. Hollanti, A. Burr, “Strong secrecy in wireless network coding systems with M-QAM modulators”, Proc. IEEE Int. Conf. Commun. China (ICCC), pp. 181-186, Oct. 2014;
C25 A. Chorti, K. Papadaki, H.V. Poor, “Optimal power allocation in block fading Gaussian channels with causal CSI and secrecy constraints”, Proc. IEEE Global Commun. (GLOBECOM), pp. 752-757, Dec. 2014;
C24 S.M. Perlaza, A. Chorti, H.V. Poor, Z. Han, “On the trade-offs between networks state knowledge and secrecy”, Proc. IEEE Int. Symp. Wireless Personal Multimedia Commun. (WPMC), pp. 1-6, Jun. 2013;
C23 A. Chorti, K. Papadaki, P. Tsakalides, H.V. Poor, “The secrecy capacity of block fading multiuser wireless networks”, Proc. IEEE Int. Conf. Adv. Tech. Commun. (ATC), pp. 247-251, Oct. 2013, (*best paper award);
C22 S.M. Perlaza, A. Chorti, H.V. Poor and Z. Han, “On the impact of network-state knowledge on the feasibility of secrecy”, Proc. IEEE Int. Symp. Inf. Theory (ISIT), pp. 2960-2964, Istanbul, Turkey, Jul. 2013;
C21 A. Chorti, S. Perlaza, Z. Han, H.V. Poor, “Physical layer security in wireless networks with passive and active eavesdroppers”, Proc. IEEE Global Commun. (GLOBECOM), Anaheim, USA, 3-7 Dec. 2012;
C20 A. Chorti, “Helping interferer physical layer security strategies for M-QAM and M-PSK systems”, Proc. IEEE CISS 2012, Princeton NJ, USA, 21-23 Mar. 2012;
C19 A. Chorti and V. Poor, ”Achievable secrecy rates in physical layer security systems with a helping interferer”, Proc. IEEE Int. Conf. Comp. Netw. Commun. (ICNC), Maui, HI, Feb. 2012;
C18 A. Chorti and V. Poor, “Faster than Nyquist interference assisted secret communication for OFDM systems, IEEE Asilomar, Pacific Grove, CA, US, 4-7 Nov. 2011, (*invited paper); C17 A. Chorti, “Masked M-QAM OFDM: Encryption of OFDM signals through faster than Nyquist
signalling”, Proc. IEEE MCECN Global Commun. (GLOBECOM), Miami, US, 6-10 Dec. 2010; C16 A. Chorti, Y. Kanaras, M. Rodrigues, I. Darwazeh, “Joint channel equalization and detection of spectrally efficient FDM signals”, Proc. IEEE Personal Indoor Multimedia Radio Commun. (PIMRC), Istanbul, Turkey, 26-29 Sep. 2010
C15 Y. Kanaras, A. Chorti, M. Rodrigues, I. Darwazeh, “A new quasi-optimal detection algorithm for a non-orthogonal spectrally efficient FDM system”, Proc. Int. Symp. Commun. Inf. Tech. (ISCIT), Incheon, Korea, 28-30 Sep. 2009;
C14 Y. Kanaras, A. Chorti, M. Rodrigues, I. Darwazeh, “An Overview of Optimal and sub-Optimal Detection Techniques for a Non Orthogonal Spectrally Efficient FDM”, Proc. LCS/NEMS, London UK, 3-4 Sep. 2009;
C13 A. Chorti, Y. Kanaras, “Masked M-QAM OFDM: A simple approach for enhancing the security of OFDM systems”, IEEE Personal Indoor Multimedia Radio Commun. (PIMRC), Tokyo, Japan, 13-16 Sep. 2009;
C12 D.T. Hristopulos, M.P. Petrakis, G. Spiliopoulos, A. Chorti, “Non-parametric estimation of geometric anisotropy from environmental sensor network measurements”, Proc. StatGIS2009, Milos, Greece, 17-19 Jun. 2009;
C11 Y. Kanaras, A. Chorti, M. Rodrigues, and I. Darwazeh, “Spectrally efficient FDM signals: bandwidth gain at the expense of receiver complexity”, IEEE Int. Conf. Commun. (ICC), Dresden, Germany, 13-17 Jun. 2009;
C10 Y. Kanaras, A. Chorti, M. Rodrigues, I. Darwazeh, “A near optimum detection for a spectrally efficient non orthogonal FDM system”, Proc. InOWo’08, Hamburg Germany, 27-28 Aug. 2008; C9 D.T. Hristopulos, A. Chorti, G. Spiliopoulos, E. Petrakis, “Systematic detection of anisotropy in spatial data obtained from environmental monitoring networks”, EGU2008, Vienna, Austria, 13-18 Apr. 2008;
C8 Y. Kanaras, A. Chorti, M. Rodrigues, I. Darwazeh, “A combined MMSE-ML detection for a Gram-Schmidt orthogonalized FDM system”, Proc. IEEE BROADNETS, London, UK, Sep. 2008;
C7 Y. Kanaras, A. Chorti, M. Rodrigues, I. Darwazeh, “Sub-optimum detection techniques for a bandwidth efficient multi-carrier communication system”, Mutli-Strand Conf., Milton, UK, 6-7 May 2008.
C6 A. Chorti, D.T. Hristopulos, “Automatic detection of spatial anisotropy in environmental data sets”, Proc. StatGIS2007, Klagenfurt, Austria, Oct. 2007.
C5 A. Moustakas, A. Chorti and D.T. Hristopulos, “Geostatistical analysis of tree size distributions in the southern Kalahari, obtained from remotely sensed data”, Proc. SPIE Europe Remote Sensing, Florence, Italy, 17-20 Sep. 2007.
C4 A. Chorti and M. Brookes, “Resolving near carrier spectral infinities due to 1/f phase noise in oscillators”, Proc. IEEE Int. Conf. Acoustics Speech Signal Process. (ICASSP), vol. 3, pp. III 1005-III 1008, Hawaii, USA, 15-18 Apr. 2007
C3 A. Chorti, D. Karatzas, N.M. White, C.J. Harris, “Intelligent sensors in software: the use of parametric models for phase noise analysis”, Proc. IEEE Int. Conf. Intelligent Sensing Inf., Bangalore, India, 15-18 Dec. 2006.
C2 A. Chorti, B. Granado, B. Denby, P. Garda, “Une architecture electronique temps reel pour les reseaux connexionnistes en physique des hautes energies”, NSI2000, Toulouse FR, May 2000. C1 A. Chorti, B. Granado, B. Denby and P. Garda, ”An electronic system for the simulation of
neural networks with real time constraints”, Proc. ACAT, Chicago, U.S., Dec. 2000.
1.3.4 Posters
P2 M. Mitev, A. Chorti, M.J. Reed, “Physical layer security in wireless networks with active eavesdroppers”, Munich Workshop on Coding and Cryptography (MWCC) 2018, (*invited poster), Germany, 10-11 April 2018;
P1 A. Chorti, “Optimal resource allocation in secure multi-carrier systems”, 1st IEEE Women’s Workshop Commun. Signal Proc., Banff, (*invited poster), Canada, 13-15 Jul. 2012.
1.3.5 In Preparation [U] / Submitted [S]
U1 M. Mitev, M. Shakiba Herfeh, A. Chorti, M.J. Reed, “Multi-factor lightweight authentication for the Internet of Things”, IEEE Trans. Inf. Forensics Security, in preparation;
U2 G. A. Nunez Segura, A. Chorti, C. Borges Magri, “Multimetric centralized and decentralized intrusion detection in software defined networks”, IEEE Internet of Things Journal, in preparation; U3 M. Bello, W. Yu, M. Pischella, A. Chorti, I. Fijalkow, L. Musavian, “A Review of DL/UL
Multiple Access Enabling Low-Latency Communications”, IEEE Access, in preparation;
U4 M. Bello, A. Chorti, I. Fijalkow, W. Yu, L. Musavian, “Performance Analysis of NOMA Uplink Networks under Statistical QoS Delay Constraints”, IEEE Trans. Communications, in preparation;
U5 N. Ferdosian, S. Skaperas, A. Chorti, L. Mamatas, “Unleashing the Potential of Flexible Nu-merology by Resolving Conflicts”, IEEE Trans. Wireless Communications, in preparation; S1 G.A. Nunez Segura, A. Chorti, C. Borges Magri, “Multimetric Online Intrusion Detection in
1.4
Recent Research Results
1.4.1 Motivation on Studying Physical Layer Security and Resource Allocation
for 5G Systems
Physical Layer Security
The goal of physical layer security (PLS) [1–3] is to make use of the properties of the physical layer – including the wireless communication medium and / or the transceiver hardware – to enable critical security aspects. In particular, PLS can be employed to provide i) node (device) authentication, ii) message authentication, iii) message confidentiality through the use of secrecy encoders, and, iv) key management and distribution solutions through symmetric secret key generation from shared randomness. Furthermore, proposals for intrusion detection and counter-jamming at PHY have recently emerged [4]; indeed these two topics emerge as important research areas in B5G systems, particularly in the industrial Internet of things (IoT) and the mmWave era.
PLS has been explicitly mentioned in the first white paper on 6G: “The strongest security protection may be achieved at the physical layer”. Importantly, it is stated as an enabling technology in the IEEE International Network Generations Roadmap 1st Edition 2019 in the Chapters on “Security” (Section 1.1 pp. 1-2) and on “Massive MIMO” (Section 4.3 pp. 8-9). The increasing interest in PLS has been stimulated by many practical needs. Notably, many critical IoT networks require ultra-low latency communications (< 1msec), e.g., in autonomous driving and vehicle to everything (V2X) applications, telemedicine and haptics. However, standard authentication often requires significant processing time. We note in passing that in the Third Generation Partnership Project (3GPP) technical report “Study on the Security of URLLC” [5], all aspects related to low latency (fast) authentication remain open and no solutions have so far been standardized. An added complication is due to hardware limitations of low-end sensors and their ineptest to execute sophisticated security protocols such as the IPSec or the DTLS.
A further challenge comes from quantum computing, which has seen significant progress after massive investment by companies such as Google, Intel and IBM to build prototypes with more than 50 qubits. In October 2019 Google published in the journal “Nature” their quantum computer experiments showing they have achieved quantum supremacy for a particular set of problems [6]. In this aspect, PLS, that relies upon information-theoretic security proofs, could resist quantum computers, unlike corresponding asymmetric key schemes relying on the (unproven) intractability in polynomial time of certain algebraic problems. Even state-of-the-art elliptic curve cryptography (ECC) schemes, that require substantially shorter keys than RSA or Diffie Hellman (DH) schemes, are still considerably more intensive computationally than their PLS counterparts and are not post-quantum.
As a result, the study of novel PLS based solution for 5G and B5G security is highly pertinent. Related proposals using physical unclonable functions [7] and secret key generation from shared randomness [8] are included in this thesis.
Resource Allocation
The roll-out of fifth-generation (5G) mobile networks and the forthcoming 6G will bring about fundamental changes in the way we communicate, access services and entertainment.With respect to the latter, the multi-fold increase in the service data rates will provide users with ultra high resolution in video-streaming, multi-media and virtual reality, offering immersive experiences. To this end, it is important for Edge content delivery infrastructures to rapidly detect and respond to changes in content popularity dynamics. For flexible and highly adaptive solutions, the capability for quick resource (re-)allocation should be driven by early (real-time) and low-complexity content popularity detection schemes. In this thesis, we study aspects of low-complexity detection of changes in video content popularity in real-time, addressed as a statistical change point (CP) detection problem [9], breaking completely new ground compared to earlier works that relied upon prediction models [10], [11].
Furthermore, novel exciting use cases were introduced in 5G in the context of ultra-reliable low latency communications (URLLC) and massive machine type communications (mMTC); the new industrial revolution, dubbed as Industry 4.0, along with emerging verticals in telemedicine, smart agriculture, etc., will bring about automation and intelligence to levels never seen before.
As 5G is required to support a large variety of services, novel solutions to enable higher resource efficiency are sought; amongst the various possible solutions, in this thesis we study non-orthogonal multiple access (NOMA) because of its advantages over conventional orthogonal multiple access (OMA) schemes in terms of spectral efficiency [12], cell-edge throughput [13], and energy efficiency [14], rendering it an attractive solution in particular for the mMTC uplink scenario.
Additionally, to account for media access control (MAC) sub-layer latency, we use the theory of the effective capacity [15], which can serve in wireless networks to provide statistical delay guarantees. The pertinence of the theory of the effective capacity as a suitable metric results from the fact that in the wireless MAC, due to small scale fading and shadowing, it is inherently impossible to provide hard delay guarantees.
In the following, a brief presentation of my principal past contributions over the last 7 years is given in reverse chronological order, to emphasize more recent results. Section 1.4.2 offers an outline of recent results in the area of resource allocation using NOMA, the theory of the effective capacity and CP analysis, while results in the area of PLS are described in Section 1.4.3.
1.4.2 Results in Resource Allocation
NOMA and Effective Capacity
Related Contributions: [J22], [J19], [C38], [C34]
In our works a flexible delay quality of service (QoS) model was employed using the theory of large deviations (G¨artner-Ellis theorem [16]) that allows defining the metric of the effective capacity (EC) in block fading additive white Gaussian noise (BF-AWGN) channels. The EC denotes the maximum constant arrival rate that can be served by a given service process, while guaranteeing a required statistical delay provisioning and is closely related to the concept of the effective bandwidth [17]. In order to capture the impact of link layer (MAC) delays in the secrecy capacity of wireless BF-AWGN channels, we introduced a novel metric, referred to as the “effective secrecy rate” (ESR); the ESR represents the maximum constant arrival rate that can be securely served (with perfect secrecy), on the condition that the required delay constraint can be statistically satisfied.
In more detail, in [J19] a novel approach was introduced to study the achievable delay-guaranteed secrecy rate, focusing on the downlink of a NOMA network with one base station, multiple single-antenna NOMA users and an eavesdropper. Two possible eavesdropping scenarios were considered; an internal, unknown, eavesdropper in a purely antagonistic network and an external eavesdropper in a network with trustworthy peers. For a purely antagonistic network with an internal eavesdropper, the only receiver with a guaranteed positive ESR was proved to be the one with the highest channel gain. The ESR in the high signal to noise ratio (SNR) regime was shown to approach a constant value irrespective of the power coefficients, while the strongest user was shown to achieve a higher ESR when it had a distinctive advantage in terms of channel gain with respect to the second strongest user. For a trustworthy NOMA network with an external eavesdropper, a lower bound and an upper bound on the ESR were proposed and investigated for an arbitrary legitimate user. For the lower bound, a closed-form expression was derived in the high SNR regime. For the upper bound, the analysis showed that if the external eavesdropper could not attain any channel state information (CSI), the legitimate NOMA user at high SNRs would always achieve positive ESR. Simulation results numerically validated the accuracy of the derived closed-form expressions and verified the analytical results given in the theorems and lemmas.
Furthermore, in [J22], [C38], [C34], we turned our attention to NOMA uplink networks. We provided performance analyses in asymptotic regimes (low and high SNR) and also proposed a novel multiple access (MA) scheme referred to as NOMA-Relevant (NOMA-R). In NOMA-R, a flexible MA
scheme is proposed based on the requirement that any user will opt for NOMA only when there is a rate gain associated. We have shown that NOMA-R outperforms both NOMA and OMA in terms of sum rates achievable in all SNR regions. Importantly, using the theory of the effective capacity we demonstrated that the NOMA-R strategy is more favorable when the target delay-bound violation probabilities are more stringent, especially for weak NOMA users.
Resource Allocation Using Change Point Analysis Related Contributions: [J17], [J20], [C32]
In [J17], [J20] and [C32] we developed novel algorithms for the real-time detection of changes in the mean and the variance of content popularity. Approaching the problem statistically, we efficiently combined off-line and on-line non-parametric CUSUM procedures. The use of non-parametric CUSUM allowed us to avoid making assumptions about the underlying statistics of the popularity of any particular content, with the additional benefit of reduced computational cost. For the detection of changes in the mean we divided the algorithm in two phases. The first phase was an extended retrospective (off-line) procedure with an improved binary segmentation step and was used to adjust on-line parameters, based on historical data of the particular video. The second phase integrated a modified trend indicator to the sequential (on-line) procedure, to reveal the direction of a detected change. We provided extensive simulations, using real data, that demonstrated the performance of the first phase of our algorithm. We also provided proof-of-concept results that highlighted the efficiency of the overall algorithm.
The approach of combining off-line and on-line CP algorithms was also employed in [J20] for the detection of changes in the variance. However, a major difference concerned the choice of the underlying test statistic, as unlike in the case of the mean, tracking changes in the variance is inherently a nonlinear estimation problem. To develop the test statistic we proposed three different approaches: i) a non-parametric approach, ii) a parametric approach using an autoregressive moving average (ARMA) model, and, iii) a parametric approach using a nonlinear generalised autoregressive conditional heteroskedasticity (GARCH) model. Our studies using synthetic data indicated that the ARMA parametric approach did not generalize well. Due to this fact, we only performed experiments on real data using the non-parametric and the GARCH approaches. We concluded that both can equally well identify large deviations in the variance and that in the general case the non-parametric approach can provide quicker detection of CPs in the datasets studied in this work. In the future, we will develop joint detectors for the mean and the variance of video content popularity.
1.4.3 Results in PLS
PLS for 5G
Related contributions: [J21], [BC3], [C37], [C33], [C27]
With the emergence of 5G low latency applications, such as haptics and V2X, low complexity and low latency security mechanisms are needed. Promising lightweight mechanisms include physical unclonable functions (PUF) and secret key generation (SKG) at the physical layer from wireless fading coefficients, as considered in [J21], [C37], [C33]. In this framework we proposed a zero-round-trip-time (0-RTT) authentication protocol combining PUF for fast authentication and generation of resumption keys using SKG. Furthermore, a novel authenticated encryption (AE) scheme using SKG and standard symmetric key block ciphers for encryption and message authentication – first proposed in [C27] – was enhanced in [J21]. Aiming at a fast PHY protocol we proposed the pipelining of the AE SKG process and the encrypted data transfer at PHY in order to reduce latency. Looking at various alternatives to implement the pipelining at PHY, we investigated a “parallel” SKG approach for multi-carrier systems (e.g., using orthogonal frequency division multiplexing (OFDM) as in LTE and 5G new radio). In the parallel approach a subset of the subcarriers was used for SKG and the rest for encrypted data transmission (using the keys generated on the subset of SKG subcarriers). The optimal solution
to the respective PHY resource allocation problem was identified under security, power and delay constraints, by formulating the subcarrier scheduling as a subset-sum 0-1 knapsack optimization [18] . A heuristic algorithm of linear complexity was proposed and shown to incur negligible loss with respect to the optimal dynamic programming solution [J21], [C37], [C33]. The proposed mechanisms, have the potential to pave the way for a new breed of latency aware PHY security protocols with an emphasis on URLLC and IoT emerging systems.
Finally, the main lines of application of PLS in B5G systems were reviewed in [BC3], starting with node authentication, moving to the information theoretic characterization of message integrity, and finally, discussing message confidentiality both in the SKG and from the wiretap channel point of view. The aim of this review was to provide a comprehensive roadmap on important relevant results by the authors and other contributors and discuss open issues on the applicability of PLS in 6G systems.
Anomaly Detection in Software Defined Networks Related contributions: [C39], [J18]
Software-defined networking (SDN) is a promising technology to overcome many challenges in wireless sensor networks (WSN), particularly with respect to flexibility and reuse. Conversely, the centralization and the planes’ separation turn SDNs vulnerable to new security threats in the general context of distributed denial of service (DDoS) attacks. State of-the-art approaches to identify DDoS do not always take into consideration restrictions in typical WSNs e.g., computational complexity and power constraints, while further performance improvement is always a target. The objective of the works in [J18], [C39] was to propose a lightweight but very efficient DDoS attack detection approach using CP analysis. Our approach was shown to have a high detection rate and linear complexity with respect to the observed time series length, rendering it suitable for WSNs. We demonstrated the performance of our detector in software-defined WSNs of 36 and 100 nodes with varying attack intensity (the number of attackers ranging from 5% to 20% of nodes).
We used CP detectors to monitor anomalies in two metrics: the data packets delivery rate and the control packets overhead. Our results showed that as the intensity of the attack increased, our approach could achieve a detection rate close to 100% and that, importantly, the type of the attack could also be inferred. As an extension of this work, we will look into distributed anomaly detection by allowing clusters of nodes to act on local early detection systems. A trade-off to be studied will concern the cluster size versus the speed of the detection while maintaining the ability to localize the source of the anomaly.
Shielding PLS Against Active Attacks
Related contributions: [BC2], [J16], [J15], [J12], [C36], [C31], [C30], [C29], [C28], [C24], [C22], [C21] SKG schemes have been shown to be vulnerable to DoS attacks in the form of jamming and to man in the middle attacks implemented as injection attacks. In [BC2] and [C36], a comprehensive study on the impact of correlated and uncorrelated jamming and injection attacks in wireless SKG systems was presented. First, two optimal signaling schemes for the legitimate users were proposed and the impact of injection attacks as well as counter-measures were investigated. Finally, it was demonstrated that the jammer should inject either correlated jamming when imperfect channel state information (CSI) regarding the main channel was at their disposal, or, uncorrelated jamming when the main channel CSI was completely unknown.
As jamming attacks represent a critical vulnerability for wireless SKG systems, in [J15], [C31], [C30], [C29], [C28] two counter-jamming approaches were investigated for SKG systems: first, the employment of energy harvesting (EH) at the legitimate nodes to turn part of the jamming power into useful communication power, and, second, the use of channel hopping or power spreading in BF-AWGN channels to reduce the impact of jamming.1 In both cases, the adversarial interaction between the pair 1We note in passing that spreading / hopping can be directly implemented with a standard inverse fast Fourier
